About the author
Boris Ginsburg
Boris Ginsburg is an applied research scientist at NVIDIA. His primary research is fast and scalable algorithms for training of neural networks.
Post by Boris Ginsburg
AI / Deep Learning
9
334
QuartzNet architecture
Develop Smaller Speech Recognition Models with NVIDIA’s NeMo Framework
By Jocelyn Huang, Yang Zhang, Boris Ginsburg and Poonam Chitale |
AI / Deep Learning
0
115
BERT Phase1 pretraining
Pretraining BERT with Layer-wise Adaptive Learning Rates
By Sharath Sreenivas, Swetha Mandava, Boris Ginsburg and Chris Forster |
AI / Deep Learning
1
43
mixed_precision_training_flow
Mixed Precision Training for NLP and Speech Recognition with OpenSeq2Seq
By Chip Huyen, Igor Gitman, Oleksii Kuchaiev, Boris Ginsburg, Vitaly Lavrukhin, Jason Li, Vahid Noroozi and Ravi Gadde |